Skill Combo - 技能组合器

• Outcome: Review
• Latest audit: May 29, 2026 at 9:31 PM UTC
• Version:
  1.0.0

Overview

This skill is a clear skill-combining coordinator, but it needs Review because it can route tasks through broad sets of installed skills without clear safety controls.

Install only if you understand that this coordinator can amplify the powers of other installed skills. Use it with trusted downstream skills, avoid combining it with skills that can post publicly, delete data, spend money, access private accounts, or create persistent tasks unless each action is explicitly confirmed, and inspect any separate runtime implementation before relying on it.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Context-Inappropriate Capability

High

• Confidence: 95% confidence
• Finding: The skill loader accepts arbitrary skill names, resolves paths, reads their content, and dynamically imports them with no allowlist, trust boundary, sandboxing, or capability restriction. In a combo/orchestration skill, this is especially dangerous because it amplifies the privileges and behaviors of every referenced skill, allowing a user to chain in unsafe or malicious skills and have them executed automatically.

Missing User Warnings

Medium

• Confidence: 89% confidence
• Finding: The README explicitly promotes combining arbitrary installed skills and lists execution modes such as sequential, parallel, and conditional execution, but it does not warn that invoked skills may perform side effects such as data access, external calls, reminders, or state-changing actions. In a skill-composition engine, this omission can cause users to combine untrusted or high-privilege skills without understanding the compounded risk, increasing the chance of unintended actions or data exposure.

VirusTotal

66/66 vendors flagged this skill as clean.